Research Opportunities – Fall 2025
ALL FALL 2025 POSITIONS ARE CLOSED
The following faculty members and PhD students are seeking research assistants this semester. All of these positions are for credit only.
PLEASE NOTE: Research Credit (GU4996 and GU4995) is only available to Undergraduates (CC, GS, EN and BC) and MAO students in the Department of Economics.
Once selected, RAs will need to register for a Research Course. Students registered for research course GU4996 will receive either 1 or 2 college credits and be charged for those chosen credits (relevant only to students who pay by the credit). GS (General Studies) students have the option of participating in a research project at no cost by instead registering for GU4995 for 1 credit, for which they will not be billed. However, in the case of GU4995, the 1 credit may not be used to fulfill the minimum credit requirement of a Columbia degree.
In both cases, students will receive a letter grade on their transcript for their work as an RA. However, in either case, research credit may not be used as a substitute for elective or seminar requirements in the major.
IMPORTANT NOTE: Students can only register for one Research project for credit. A second project can be undertaken for the experience only, and without credit.
If interested in an RA position, please do the following:
1) Contact the researcher directly at the email address provided, and include a copy of your Columbia transcript (unofficial is ok) as well as your CV/resume.
If you are selected as an RA by the researcher, continue with the additional steps below:
2) Contact Cassandra Kessler at c.kessler@columbia.edu to let her know whom you will be working with, and cc the researcher on your email.
3) You will then be sent a link to an RA form to fill out.
UG RA Open Positions Fall 2025
Seyhan Erden (Faculty)
seyhan.erden@columbia.edu
1. Conducting Experiments for the Metrics Mentor Project
Metrics Mentor is an interactive graphical user-paced online platform. We do experiments in the undergrad econometrics classes to see its effectiveness.
We need research assistants to help with these experiments and piling the data from the experimental and control groups. Stata knowledge is necessary.
2. Evaluating the Evaluations / Grade Inflation
We use registrar’s data on class, student and instructor data to understand evaluations and grade inflation (co-authored with Waseem Noor)
RA must be an MA student in their 2nd or 3rd semester (so please apply towards the end of the Fall semester)
3. Attitudes toward Econometrics Courses
We do a survey at the beginning and at the end of each semester to collect data about attitudes towards the econometric class
RA must be an MA student in their 2nd or 3rd semester. Must know how to clean data and must be able to use Stata
Zihao Li (PhD Candidate)
zl3366@columbia.edu
Microeconomics Research
This is an open ended project. The student will read economics theory papers with me together and work on some serious research projects. It will be beneficial for students who aim to apply to economics PhD programs and interested in Game Theory. My personal background can be found at: https://www.zihaoliecon.com/
Very self-motivated. Extremely strong mathematical background.
Fanyu Wang
fw2397@columbia.edu
1. Forecasting the Truth: Measuring the Accuracy and Impact of Public Weather Forecasts in China
How reliable is the weather forecast that dictates our daily decisions? As climate change intensifies, leading to more frequent and severe extreme weather events, the accuracy of public weather forecast becomes a critical matter of public safety and health. This project involves building a novel dataset by collecting in real time forecast and alert data from the China Meteorological Administration (CMA). The core immediate goal is to create a high-quality dataset and understand the institutional workflow of weather forecast production in China. This work lays the essential groundwork for future research into how forecast quality impacts health outcomes, such as mortality rates during extreme heat or cold.
We are looking for a student to support the following key tasks. (1) Writing up codes for scraping in real time weather forecast and alert data from official government websites. (2) Cleaning, organizing, and managing the collected data. (3) Researching and summarizing the institutional background and workflow of how forecasts are produced and issued by different levels of governments. (4) Perform exploratory data analysis.
Proficiency in Python or R is required. Strong Chinese reading skills are essential for navigating government websites. Experience with web scraping is a strong plus, as well as interest in environmental/health economics. General programming and computing background is preferred.
2. Homeowner Insurance in the AI Age
How does the adoption of artificial intelligence–powered risk models transform insurance markets and affect households? In recent years, insurers have increasingly relied on granular property-level risk assessments (such as CoreLogic’s wildfire and flood models) to classify and price risk. These tools have the potential to improve efficiency by aligning premiums with true risk, but they also raise concerns about affordability, equity, and market stability. At the same time, consumers may increasingly interact with AI applications that provide easier access to personalized risk evaluations, premium quotes, or decision-support tools. This project seeks to explore how AI applications affect both insurers and consumers, and how these developments interact with insurance regulation and household welfare, with a focus on the U.S. homeowner insurance market. The immediate goal is to build institutional knowledge and collect preliminary data that will serve as the foundation for understanding the economics of AI in insurance.
The RA will play a central role in both background research and early-stage empirical work. Key tasks include: (1) Conducting institutional and policy research on the practice and the regulation of homeowner insurance markets, including new development related to AI applications. (2) Collecting and cleaning publicly available data from state insurance departments (e.g., California Department of Insurance filings, premium and cancellation data). (3) Performing exploratory data analysis on premium trends, market exits, and geographic variation in risk exposure. (3) (Optional) Assisting with preliminary theoretical modeling, for those with interest in applied micro theory and insurance economics.
I am looking for an enthusiastic student with interest in insurance markets and the economics of AI applications. Ability to synthesize policy documents and summarize institutional settings clearly is required. Proficiency in Python or R for data scraping, cleaning, and analysis is preferred. Familiarity with econometrics or microeconomic theory is a plus, but not required.
Ruixue Li
rl3320@columbia.ed
1. Evaluating the Health Impact of Three North Afforestation Project in China
The Three North Shelterbelt Program (三北防护林) is the world’s most ambitious afforestation project. This project aims to identify the causal effect of afforestation on reducing dust storms and hence improving health outcomes of the downwind locations.
The RA will help with data collection, data cleaning, and literature review, depending on the RA’s interests and skills. Fluency in Chinese is necessary, programming skills in Python or familiarity with satellite imagery is preferred.
2. Early-Stage Projects on Political Economy
I have a few early-stage projects in political economy that require gathering and processing information (texts, and potentially images) on social media / news outlets / other sources. I’m looking for both RA and collaborators on some projects.
Intermediate/advanced programming skills and good coding etiquette are necessary (experience with web scraping, text/image analysis, LLM, is highly preferred). Fluency in major Asian/European languages would be useful as well.
Tianling Luo (PhD Candidate)
tl3078@columbia.edu
Learning and Mental Models
The project is related to learning and the formation of mental models. We are interested in studying how individuals form and update their mental models, both theoretically and experimentally.
I am looking for assistance with a literature review (in both psychology and machine learning) as well as with setting up related experiments. A background in computer science would be particularly valuable for this work.
Brett House (Faculty)
beh2121@columbia.edu
CBS Core & Elective Course Material Updates – CLOSED
As the Core Macro Course Coordinator at CBS, prepare, update, and revise our teaching materials for the entire macro group and the intern would assist me in completing this year’s additions and edits. The would also assist in compiling, formatting, and visualizing course materials for our electives. They may also assist in research and writing for new CBS cases.
Facility with Python, Stata, FRED, Bloomberg, DBnomics, and related data platforms would be useful.
Catalina Gomez (PhD Candidate)
cmg2267@columbia.edu
Public Economics of Health and Social Services in Brazil – CLOSED
I am looking for Research Assistants to support a range of projects in public economics, with a particular focus on the provision of health systems and social programs. These projects use tools from applied microeconomics to study how government policies shape individual outcomes and institutional performance. One current project investigates how the expansion of urgent care centers in Brazil’s public health system affects hospital use and overall system efficiency, using large-scale administrative data. Other projects may explore related topics such as the impact of waiting times in healthcare, or the targeting of social programs.
Potential tasks include literature reviews, background research on institutional features, documentation of policy or legal reforms, and assistance with data preparation or descriptive analysis. Students interested in learning R are welcome. A strong interest in public policy and applied research is more important than prior technical training.
Maya Norman
man2185@columbia.edu
The Priorities of US Federal Land Management
The U.S. federal government manages nearly half of all land in the Western United States. By law, most of these lands must be governed under a sustained-yield, multiple-use mandate. This means they are to be managed in ways that preserve their value for future generations and balance competing uses. For example, oil and gas development and ecosystem conservation should be prioritized equally. This project investigates the extent to which federal land management practices align with this legal mandate.
Tasks: (i) Read and synthesize bureau of land management resource management plans (dense environmental planning documents). (ii) Engineer LLM prompts to extract specific information from these documents. Skills: (i) excellent English language skills are a must, (ii) intermediate python programming skills.
Zhihui Wang (PhD Candidate)
zw2957@columbia.edu
1. Altruism in Networks – CLOSED
This project studies how altruism shapes network formation and strategic behavior. When altruistic preferences are determined ex ante—such as when individuals know beforehand whether they hold universalistic (caring equally about everyone) or particularistic (favoring a specific group) attitudes—how does this influence their decisions to form friendships, select trading partners, or hire employees?
Undergraduate RA tasks will include reading the literature, brainstorming models, and empirical strategies.
2. Hierarchical Hold-up – CLOSED
Governments are typically hierarchical, comprising both a central government and multiple
local governments. Local governments, being closer to firms, are more directly involved in negotiation and implementation. The central government, by contrast, often serves as the ultimate guarantor of legal and policy enforcement. This project studies a dynamic hold-up problem involving three actors: a firm, a local government, and a central government. In each period, the firm decides whether to invest; the local government sets a tax rate; and the central government chooses whether to enforce tax bounds when violations occur.
Undergraduate RA tasks will include reading the literature, brainstorming models, and empirical strategies.
Waldo Ojeda (Faculty)
wo2198@columbia.edu
1. Estimating the Housing Effects of Mexican Repatriations in the 1930s
This project will study the impact of Mexican repatriations in the 1930s on the housing market. Due to the Great Depression forcing Americans into unemployment, an estimated 400,000 Mexican and Mexican-Americans living in the United States were coerced or forced into deportation to Mexico. Using the full-count Census data, we will estimate the impact of Mexican repatriations on home prices, rents and owner-renter transitions for Americans that continued to live in the United States.
Undergraduate RA tasks and skills: R, Data Processing, Data Analysis, Python, Stata, Econometrics, GitHub
2. Corporate Discount Rates and Management Departures
This project will study the impact of corporate discount rates on corporate investments. To estimate a causal effect, we will instrument for corporate discount rates with sudden corporate management departures. We will use Compustat data to understand corporate financial behavior and merge it with management departure and corporate discount rates from prior research.
Undergraduate RA tasks and skills: R, Data Processing, Data Analysis, Python, Stata, Econometrics, GitHub
Eshaan Patel (PhD Candidate)
eshaan.patel@columbia.edu
1. Voter Responsiveness to a Crisis: Evidence from the Texas Winter Storm
Who do voters blame for a crisis? In 2021 the Texas Winter Storm created unprecedented widespread power outages across the state. Leveraging variation in the severity of these outages, this project seeks to observe how voting behavior (e.g. incumbency support, voter turnout, 3rd party support) changed as a result of exposure to this crisis.
This project allows RAs to engage deeply with all parts of the research process, especially data analysis. Qualified RAs will be self-driven and excited to conduct independent research. Econometrics coursework and knowledge of difference-in-differences research design is required. Experience with data analysis and coding in STATA (or a similar language) is a plus but not required.
2. Power, Targeting, and Firm Growth
There are dramatic differences in institutions and power across countries. In a context with weak institutions, are powerful firms able to use the political sector to target their less powerful competitors? This project seeks to identify the implications of this channel using data from Latin America to see if it can explain aggregate data on firm size and growth. RAs will have the opportunity to develop skills in applied economic research.
RAs will work with geo-coded firm address data and use GIS to make maps and create distance measures between firms. No prior experience required, but experience with ArcGIS/QGIS, Python, and basic knowledge of Spanish is helpful. Qualified RAs will demonstrate ability to pick up new coding skills.
3. The Nature of Firm Lobbying
The lobbying industry in the United States is large, and the origin of these lobbying revenues are often firms and industry-associations. This project aims to use recent advances in natural language processing to classify lobbying activities into those related to general, productivity-enhancing policy versus specific, rent-seeking activities. RAs will have the opportunity to help build a novel dataset on lobbying activity.
RAs will extract congressional bill level data (bill text, legislative history) from a congressional database using an API. Qualified RAs will have programming experience. Prior experience working with APIs is helpful but not required.
Joshua Thomas (PhD Candidate)
joshua.k.thomas@columbia.edu
Community Investments in Canadian First Nations Communities
This project analyzes financial structures and community investment patterns among Canadian First Nations communities using detailed financial statement data from 2013–2025. We aim to identify factors that influence community spending behaviors and measure the impact of investments on various economic and social outcomes. This work will contribute to understanding how self-governance and fiscal autonomy affect community development and welfare.
The research assistant(s) will work on numerous tasks related to extracting and digitizing financial data from PDF documents, including both digital files and lower-quality scans. Tasks may include assisting with the development of a data collection plan and extracting several data types, such as writing Python-based OCR extraction programs, utilizing external digital tools, and performing manual data entry.
Strong programming skills in Python are essential, ideally with some experience in document processing and extraction. However, applicants with limited programming skills who are interested in the project and willing to perform manual data entry and quality checks will also be considered. An interest in public economics, development economics, and topics surrounding Indigenous communities is an asset.
Donald Davis (Faculty)
drdavis@columbia.edu
Urban and Spatial Economics
Assist with a variety of projects on urban and spatial economics. My current research deals with racial segregation, knowledge spillovers in cities, and how the latter affects wage dynamics and inequality across cities. I am also looking at opportunities to apply AI in research and teaching at both graduate and undergraduate levels.
Tasks depend on skills of the RA. Early tasks will involve assembling research articles on a variety of spatial topics. Later tasks could include coding in R for analysis and mapping of spatial data. Familiarity with AI platform(s) helpful.
Stephan Thies
stephan.thies@columbia.edu
Public Transportation Pricing and Labor Monopsony Power
This project investigates how public transit pricing impacts market power of firms in the labor market. We study the introduction of a nationwide public transit pass at heavily subsidized prices in Germany to understand if better market accessibility improves employment outcomes. This might be especially relevant for workers in rural areas, workers in their early career, and other employees who face significant costs from buying a car. This is a topic at the forefront of research in urban and labor economics and results will inform policy making in transportation.
We are looking for an enthusiastic student to support us in the collection of public transit price data in Germany. This includes scraping of the website of the German national rail operator as well as extraction of more detailed information from ticket pdf leaflets of regional operators. The project includes the use of state of the art AI models for image processing and information extraction.
Decent German skills are required, coding experience in Python or R (or a willingness to learn either) is very desirable, as is an interest in urban and labor economics.
Dongcheng Yang (PhD Candidate)
dy2426@columbia.edu
Credit Guarantee and Financial Misallocation – CLOSED
This project studies how government-backed credit guarantees shape bank–firm relationships and firms’ access to debt. While the policy aims to ease financing constraints, it may also lead to inefficient allocation of credit across firms, industries, and regions. Using a combination of econometric and modeling approaches—including instrumental variables, difference-in-differences, and general equilibrium analysis—the study aims to quantify the welfare implications of credit guarantee programs.
RA will primarily assist with data cleaning and data analysis. The ideal candidate should be proficient in statistical software such as Stata and Python. Experience with ArcGIS or other GIS tools is a strong plus, as the project involves spatial analysis of policy impacts. Japanese language skills are preferred but not required.
Ankit Bhutani (PhD Candidate)
ab4462@columbia.edu
1. Impact of Delayed IPOs – CLOSED
The growth of venture capital (VC) and private equity (PE) funding has allowed many startups to delay going public. This project examines how this shift affects capital allocation and the performance of already listed peer firms. RAs working on this project will gain hands-on experience with data analysis and work with datasets widely used in academic research and the finance industry.
Skills required: Working knowledge of at least one programming language – preferably Python, R, Julia, or Stata. Preference to students who have taken at least one econometrics course.
2. Environment Related Corporate Disclosures – CLOSED
Many large U.S. and global companies began publishing standalone sustainability or CSR reports during the early 2010s. These reports often contain disclosures of greenhouse gas (GHG) emissions, including Scope 1 (direct) and Scope 2 (indirect, electricity-related) emissions. This project investigates firms’ Scope 1 and Scope 2 emissions disclosure behavior using computational text analysis.
Tasks: (i) Find and collect sustainability reports for U.S. public firms. (ii) Use Python to parse and search the reports. (iii) Write scripts to detect and code Scope 1 and Scope 2 disclosure. (iv) Record and organize outputs systematically in a spreadsheet.
Skills: (i) beginner-to-intermediate Python programming (PDF/text parsing, regex, data cleaning), (ii) basic Excel/Google Sheets skills, (iii) attention to detail in coding and validation.
Kamelia Stavreva (PhD Candidate)
kes2220@columbia.edu
Skin Tone Inequality – CLOSED
This project uses historical data to study gaps across a variety of outcomes (such as income, migration, homeownership, and occupation) between individuals with different skin colors. The project currently focuses on differences for Black people and asks why these differences emerge, how they impact their lives, how these impacts transmit over generations, and how they have changed over time. The project also examines how these skin color differences relate to broader changes in racial inequality over time. A research assistant will help analyze and work through a large set of historical data from the late 1800s to the present day.
The RA should have strong skills in data analysis and data cleaning using STATA, R, or Python. Familiarity with GitHub, Overleaf, SQL, LaTeX, and cloud platforms (e.g., AWS ) is a plus.
Sahila Kudalkar
s.kudalkar@columbia.edu
Bureaucratic Incentives for Environmental Protection
This project examines how political influence shapes bureaucrats’ incentives in granting environmental clearances for mining and infrastructure projects in India. Using 10 years of administrative data on bureaucrat postings and project approvals, the project examines whether officer performance on the environment affects future transfers and promotions.
The RA will assist with data cleaning, processing, and analysis. This position provides hands-on experience with real-world policy data and quantitative research methods — skills that are highly transferable to future research, policy work, and graduate study.
The RA should have the following skills: Basic coding in R or Python; basic familiarity with GIS. Knowledge of Hindi is useful but not required. Curiosity about political economy and the environment is a plus.
Zikai Xu (PhD Candidate)
zx2306@columbia.edu
Cheap Talk in Search Markets – CLOSED
We study a model of search market where a consumer can only acquire information via cheap talk from a visited seller. We focus on equilibria where there are infinite, symmetric sellers. Our goal is to characterize the set of equilibrium payoffs and its boundary in general cases.
The RA will be participating in the following tasks: 1. Proofread the paper 2. regularly discuss the related literature.
Junho Choi
jc5341@columbia.edu
1. AI Training, Energy Use, and Environmental Consequences
This project studies the energy and emissions impacts of large-scale AI training, with a focus on Google’s data centers in regions such as Iowa and Oklahoma. We investigate whether intensive AI training draws on fossil fuel versus cleaner energy sources, and how this affects emissions like SO2 and NOx. The analysis will also extend to the potential health and labor market consequences of fossil-fuel-driven energy use during these training periods.
RAs will collect and organize data on energy production, emissions, and AI training timelines, as well as assist in cleaning, merging, and analyzing large datasets. Tasks may include web scraping, exploratory data analysis, and documenting results. Skills desired include proficiency in Python or R (required), basic understanding of econometric/statistical methods (required), and experience with data gathering or scraping (a strong plus). A solid programming/computing background is preferred.
2. AI, Energy, and Productivity: Understanding the Impact of AI Outages on Workers and Firms
This project examines how artificial intelligence (AI) affects worker and firm productivity, with a focus on the role of AI reliability and energy use. By analyzing periods of OpenAI service outages, we can study how disruptions influence business performances. We also hope to explore the broader energy and emissions impacts of AI adoption, bridging perspectives from energy economics and business economics.
The research assistant will help gather and clean data on AI outages, productivity measures, and energy use, and assist with building datasets suitable for econometric analysis. Tasks may include web scraping, organizing large datasets, performing exploratory analysis via econometric methods, and documenting workflows.
Proficiency in either Python or R (required); Basic understanding of econometric/statistical methods (required); Experience with data gathering or scraping (strong plus); General programming and computing background (preferred).
Hannah Solheim (PhD Candidate)
h.solheim@columbia.edu
A Study about Social Media Algorithms – CLOSED
My co-authors and I are looking for up to three RAs to assist with projects broadly related to behavioral/experimental and applied microeconomics. Topics include how social media content shapes self-perception and discrimination against others, how medical innovation affects health disparities, and how intermediaries affect the search for talent in labor markets.
The RA will have a variety of tasks, which may include: experiment prep, literature reviews, and data analysis. Knowledge of STATA or Python is a plus, but not required. If you do have coding experience, please include a sample of code with your application.
Pablo Mones (PhD Candidate)
pm3257@columbia.edu
Counting and Characterizing Compliers and Noncompliers
Counting and characterizing compliers and noncompliers in instrumental variables (IV) settings under the local average treatment effect (LATE) framework is widely used to understand generalizability, treatment take-up, and variation in treatment effects across different instruments. Several identification methods have been proposed to characterize compliers; however, we find that when accounting for overlooked identification assumptions, Abadie’s kappa-weighting scheme is more general and relies on fewer assumptions, though it is more challenging to estimate. Building on this result, we extend the kappa-weighting scheme to account for noncompliers and propose a general approach for counting and characterizing both groups. Finally, we introduce a partial identification strategy to assess how relaxing monotonicity—an increasingly debated assumption—affects the counting and characterization of compliers and noncompliers.
Tasks:
We aim to construct a package on R containing functions that implement the results derived in the paper. The chosen RA will mostly work on this by creating functions that perform the different estimation strategies discussed in the paper.
Skills: We are looking for the following skills:
(i) Advanced knowledge of R and Python will be required.
(ii) The student should have previous knowledge of econometrics.
(iii) Students with a strong quantitative profile will be preferred. Having taken courses in computer science, mathematics or statistics will be specially valued.
(iv) Knowledge of other statistical software (Stata, Matlab, etc) will also be valued.
Waseem Noor (Faculty)
wn4@columbia.edu
Global Economy Course Update – CLOSED
The Global Economy course will be taught in Spring 2025. This project will update the weekly readings for the course, and suggest ideas for the exam and midterm. The student could also be a Teaching Assistant for the spring semester.
Good research skills, ability to work with Excel, PowerPoint and Word, and an interest in international economic issues a must. Preference for students who have taken international trade or global economy (with Prof Noor). Avid readers of the Economist and other periodicals and newspapers are encouraged to apply.
